I've heard some people argue that the Ninth Circuit's holding (in Nordyke v. King) that the Second Amendment is incorporated against the states was "dictum" -- and thus not really legally binding, even on district courts in the Ninth Circuit and other panels of the Ninth Circuit -- because the court went on to hold that the Second Amendment (even as incorporated) doesn't preclude the ordinance involved there, which was a restriction on gun possession on county property. (Likewise, people made the same argument about the Fifth Circuit's pre-Heller holding in United States v. Emerson that the Second Amendment secures an individual right, but doesn't preclude the particular statute involved in that case.) The conclusion that the Second Amendment binds state and local governments, the argument goes, isn't really necessary to the result because the court could have reached the same conclusion by simply assuming without deciding that the Second Amendment is incorporated. Therefore, the incorporation conclusion is mere "dictum."
I think this is a mistaken conception of dictum, but much scholarly (and some judicial) ink has been spilled on the theoretical question of how dictum should be defined. Instead, I just wanted to repeat something I said earlier (in the Emerson context): If Nordyke v. King is dictum, then some extremely important Supreme Court rulings are dictum, even though they have to my knowledge never been treated as such. Here are three:
1. Jackson v. Virginia (1979). Jackson claimed that there was insufficient evidence to convict him in his state trial, and that she had a federal Due Process Clause right to have her conviction reversed on these grounds. To resolve this question, the Court first had to figure out whether the Due Process Clause secured such a right (a matter that was quite unresolved at the time, and that was contested by the Virginia prosecutors). The Court addressed this in some detail, and concluded that the Due Process Clause secured such a right. But it concluded that the right was limited in scope: The Due Process Clause is violated only if "upon the record evidence adduced at the trial no rational trier of fact could have found proof of guilt beyond a reasonable doubt." This condition wasn't satisfied in this case, so Jackson lost.
2. Strickland v. Washington (1984). Strickland claimed that his lawyer provided ineffective legal assistance, and that this violated the Sixth Amendment right to counsel. To resolve this question, the Court first had to figure out whether the Sixth Amendment secured such a right (likewise, a matter that was quite unresolved at the time). The Court addressed this in some detail, and concluded that the Sixth Amendment secured such a right. But it concluded that the right was limited in scope: The Sixth Amendment is violated only if "counsel made errors so serious that counsel was not functioning as the 'counsel' guaranteed the defendant by the Sixth Amendment," and "[j]udicial scrutiny of counsel's performance must be highly deferential . . . [--] a court must indulge a strong presumption that counsel's conduct falls within the wide range of reasonable professional assistance; that is, the defendant must overcome the presumption that, under the circumstances, the challenged action 'might be considered sound trial strategy.'" This condition wasn't satisfied in this case, so Strickland lost.
3. Yesterday's Safford Unified School Dist. v. Redding. Redding claimed that she was entitled to damages for violation of the Fourth Amendment, because a near-strip-search at school violated her Fourth Amendment rights, and those rights were well enough established that the defendants didn't have qualified immunity against damages. To resolve this question, the Court first had to figure out whether her Fourth Amendment rights were indeed violated. The Court addressed this in considerable detail, and concluded that there was indeed a Fourth Amendment violation. But it concluded that the rule wasn't well enough established, so Redding lost. (If you think that for some reasons qualified immunity cases are a different category, even after reaching the merits first became optional with the demise of Saucier v. Katz, then ignore this example and focus on the three others.)
4. Everson v. Board of Ed. (1947), an incorporation case. Everson claimed that a particular government policy violated the Fourteenth Amendment, because the Fourteenth Amendment incorporated the Establishment Clause against the states, and the policy violated this incorporated Establishment Clause. To resolve this question, the Court first had to figure out whether the Fourteenth Amendment did indeed incorporate the Establishment Clause. The Court addressed this fairly briefly, and concluded that the Fourteenth Amendment did incorporate the Establishment Clause. But it concluded that the Establishment Clause didn't bar all evenhanded government benefit programs that covered religious schools as well as secular schools. The threshold for violation of the Establishment Clause wasn't met in this case, so Everson lost.
5. So now we have Nordyke, another incorporation case. Nordyke claims that a particular government policy violated the Fourteenth Amendment, because the Fourteenth Amendment incorporated the Second Amendment against the states, and the policy violated this incorporated Second Amendment. To resolve this question the Ninth Circuit panel first had to figure out whether the Fourteenth Amendment did indeed incorporate the Second Amendment. The court addressed this in considerable detail, and concluded that the Fourteenth Amendment did incorporate the Second Amendment. But it concluded that the Second Amendment gives the government a fairly free hand to restrict gun possession on government property. The threshold for violation of the Second Amendment wasn't met in this case, so Nordyke lost.
My point is simply that all five of these examples (perhaps with the exception of item 3, if you think qualified immunity cases are somehow different) are on par with each other. If you think that Nordyke's incorporation discussion is dictum and not strictly binding, the same would have been true of all the other cases, at least until the first decisions actually finding liability under those cases were reached. But, as I said, to my knowledge no-one has treated the core legal reasoning of Jackson, Strickland, and Everson as dictum. Nor should they do the same as to Nordyke.